Shaozhen
Jing†
ade,
Xiaolei
Wu†
ade,
Daniel Shiu-Hin
Chan†
c,
Sang-Cuo
Nao
b,
Jianxiong
Du
a,
Chun-Yuen
Wong
c,
Jing
Wang
*ade,
Chung-Hang
Leung
*bfgh and
Wanhe
Wang
*ade
aXi'an Key Laboratory of Stem Cell and Regenerative Medicine, Institute of Medical Research, Northwestern Polytechnical University, 127 West Youyi Road, Xi'an, Shaanxi 710072, China. E-mail: whwang0206@nwpu.com; jwang0321@nwpu.edu.cn
bState Key Laboratory of Quality Research in Chinese Medicine, Institute of Chinese Medical Sciences, University of Macau, Taipa, Macau. E-mail: duncanleung@um.edu.mo
cDepartment of Chemistry, City University of Hong Kong, Tat Chee Avenue, Kowloon, SAR, Hong Kong
dNorthwestern Polytechnical University Chongqing Technology Innovation Center, Chongqing 400000, P. R. China
eResearch & Development Institute of Northwestern Polytechnical University in Shenzhen, 45 South Gaoxin Road, Shenzhen 518057, China
fDepartment of Biomedical Sciences, Faculty of Health Sciences, University of Macau, Taipa, Macau
gMacao Centre for Research and Development in Chinese Medicine, University of Macau, Taipa, Macau
hMoE Frontiers Science Centre for Precision Oncology, University of Macau, Taipa, Macau
First published on 8th May 2024
Iridium(III) complexes are alternative bioimaging probes due to their tunable photophysical properties, but are limited by poor cell penetrability and high cytotoxicity. Recently, iridium(III)–peptide bioconjugates have received significant attention as bifunctional molecules in bioanalytical and biomedical fields. Conjugation to peptides endows iridium(III) complexes with specificity, potentially overcoming the side effects and drug resistance of metallodrugs, whilst enhancing cellular uptake due to the improved cell penetrability, low cytotoxicity and targetability of peptides. In this review, we briefly introduce the interactions between iridium(III) complexes and amino acids/peptides, including coordination to amino acids and detection and/or inhibition of peptides. We describe imaging applications of iridium(III)–peptide bioconjugates, involving direct coordination of functional peptides or ligand modification, for targeted imaging. Next, we present therapeutic and theranostic applications of iridium(III)–peptide bioconjugates through targeting of DNA and proteins. Finally, we outline the challenges and future opportunities in the development of iridium(III)–peptide bioconjugates for precision medicine.
Xiaolei Wu is a master's student in the Institute of Medical Research at the Northwestern Polytechnical University, and her research interests focus on the development of luminescent sensing platforms for disease-related analytes. |
Jianxiong Du is a master's student in the Jiangxi Provincial Key Laboratory of Functional Molecular Materials Chemistry at the Jiangxi University of Science and Technology, and his research interests focus on the development of luminescent sensing platforms for disease-related analytes. |
Jing Wang obtained her PhD degree at Sun Yat-sen University, then undertook her postdoctoral research in the group of Prof. Dennis Lo at the Chinese University of Hong Kong. She is currently working as an Associate Professor at the Institute of Medical Research at the Northwestern Polytechnical University. Her research interests include the development of visual sensing platforms based on nucleic acid amplification and gold nanoparticles for non-invasive disease diagnosis. |
Wanhe Wang obtained his MSc degree at Jilin University, then completed his PhD degree at the Hong Kong Baptist University. He is currently working as Associate Professor at the Institute of Medical Research at the Northwestern Polytechnical University. His research interests include the development of luminescent sensing platforms for disease-related biomolecules and metallodrugs. |
Peptides are composed of multiple amino acids joined by amide bonds, and show versatile functions in various aspects of biological functions, such as cancer cell targetability and subcellular localization.16,17 Considerable efforts have been made towards the development of peptide molecules with precisely tailored biological and structural properties over the past decades.18–21 Meanwhile, peptides have also received intensive attention for analytical and therapeutic applications in the past two decades.22–25 Peptides occupy an intermediate chemical space between small organic molecules and large biologics for drug design and discovery.26,27 In drug discovery, peptide-based molecules offer advantages of high biological activity, high specificity, and low toxicity.28 However, peptides can suffer from low stability and lack of oral bioavailability, limiting their pharmaceutical potential.29 Hence, peptide conjugation has become a popular strategy for the development of clinically relevant therapeutic agents.30,31
Compared with other transition metal complexes, iridium(III) complexes have distinct luminescence properties that make them attractive as bioimaging probes.5,7,32,33 For example, iridium(III) complexes show intense phosphorescence at room temperature with a long emission lifetime, while complexes of rhodium(III), which is also in group 14, only give measurable emission at low temperatures, consistent with stronger spin–orbit coupling of iridium(III) relative to rhodium(III).34 Moreover, the emission color of iridium(III) polypyridine complexes can be readily tuned from bluish-green to near-infrared (NIR) through the use of different cyclometalated/ancillary ligands, while emissive ruthenium(II) polypyridine complexes are mostly confined to the orange to red region.35,36 Furthermore, most iridium(III) complexes exhibit high intrinsic mitochondrial specificity due to their cationic and lipophilic properties, and can be modified to target other organelles with high specificity, while ruthenium(II) complexes generally show lower lipophilicity and poor membrane permeability.37,38
However, iridium(III) complexes can sometimes display poor cell penetrability and high cytotoxicity, preventing their further application in biological systems.39 On the other hand, peptides have low cytotoxicity, good targetability and/or high cell penetrability. Many functional peptides have been applied in peptide–drug conjugates (PDCs) to potentially improve drug targeting, reduce side effects in other cells, and improve drug bioavailability.29,40 Additionally, certain peptide sequences, both natural and synthetic, have specific receptors that are overexpressed in cancer cells compared with normal cells, which can be exploited for targeted drug delivery.41,42 Therefore, iridium(III)–peptide bioconjugates could successfully combine the advantages of iridium(III) complexes and peptides, while overcoming the cytotoxicity of the iridium(III) complex.
Furthermore, the introduction of cationic and lipophilic iridium(III) complex can overcome another drawback of peptides: their low metabolic stability and poor oral bioavailability.43,44 In the field of peptide chemistry, certain strategies for improving the stability of peptides have been explored, such as cyclization, increasing the steric bulk of side chains, D-amino acids, N-acetylation and C-amidation (N-methylation, modification in the end of peptide by polyethylene glycol) or adding fatty chains.45,46 Thus, conjugation with iridium(III) complexes could provide an alternative method to improve peptide stability, complementing existing techniques.45,47–49
A few strategies are available for synthesizing iridium(III)–peptide conjugates. Generally, peptides can be first synthesized either through solid-phase peptide synthesis or solution-phase peptide synthesis. Subsequently, peptides can be conjugated to iridium(III) complexes by amide bond formation, click chemistry, or coordination to the side chain of amino acid residues (e.g. imidazole in histidine).50 Amide bond formation is a classical reaction that is widely used in all areas of organic chemistry.51 Click chemistry is a relatively newer strategy that can be used for preparing more structurally challenging bioconjugates and for efficiently establishing a library of bioconjugates.52,53 For some peptides that have specific amino acids such as histidine, the coordination strategy could be preferable for rigidifying the peptide for better affinity.
A variety of iridium(III) complexes functionalized with bioactive peptides have been extensively applied in cell imaging, drug discovery and other areas.54–56 However, although a few reviews describing metal–peptide bioconjugates have been published,44,51,57–61 specific reviews on the bioanalytical and biomedical applications of iridium(III)–peptide bioconjugates are scarce. In this review, we summarize the interactions between iridium(III) complexes and peptides and the emerging applications of iridium(III)–peptide bioconjugates in biomedical fields (Scheme 1), highlighting their widespread application for targeted luminescent imaging and as therapeutic agents. However, we exclude examples of using peptides as a linker to construct iridium(III) complexes, as the peptide moiety only plays a minor role in such complexes.62 We also outline the challenges and future opportunities in the development of iridium(III)–peptide bioconjugates and applications for precision medicine.
![]() | ||
Scheme 1 Overview of luminescent iridium(III)–peptide bioconjugates for bioanalytical and biomedical applications. |
Alzheimer's disease (AD) is a prevalent neurodegenerative disorder, with over 51.6 million individuals with AD-related dementias worldwide in 2019 and nearly 9.83 million individuals with AD in China in 2020.64 One main hallmark of AD is the aggregation of amyloid-β (Aβ) peptides in the brain of AD patients.65,66 Aβ is a native metal-binding peptide with a typical N-terminal metal-binding sequence including three histidines,67 and copper dyshomeostasis was also observed in AD patients.68 These studies inspired the application of transition metal complexes for interacting with amino acids and Aβ. In 2008, a landmark study by Ma's group applied a solvento iridium(III) complex 1 for the detection of histidine and histidine-rich proteins.69 The labile solvento ligands of complex 1 could be replaced by histidine to form iridium(III)–peptide conjugates, similar to the interaction between Aβ and copper ions. After this, the same group further synthesized two solvento iridium(III) complexes 2 and 3 with water (H2O) ligands as luminescent probes for Aβ1–40 peptide and as inhibitors of amyloid fibrillogenesis (Fig. 1), which detected and inhibited Aβ1–40 through the replacement of labile H2O ligands by the imidazole N-donor moiety of histidine residues of the Aβ1–40 peptide to form a conjugate.70 Moreover, other transition metal complexes, such as copper(II), platinum(II), and ruthenium(II) complexes, have also been found to inhibit the aggregation of Aβ peptides.71
Ma's group further reported that iridium(III) complexes with diverse ligands could inhibit Aβ1–40 peptides via non-covalent interactions.72 They synthesized a series of iridium(III) complexes based on previous C^N ligands, and demonstrated that the interactions between iridium(III) complexes and peptides were not confined to solvento iridium(III) complexes. The top candidate complex 4 was identified as a dual imaging and modulating probe of Aβ for the treatment of AD.
Cyclometalated iridium(III) complexes can also be used as photosensitizers due to their ability to generate reactive oxygen species (ROS) under light irradiation.73–75 ROS could be produced through an electron transfer (˙OH, O2˙−)-based type I pathway or an energy transfer mechanism (1O2)-based type II pathway.76–78 In 2016, Lim's group reported that complex 5 with a dimethylamino group on the N^N ligand could modulate amyloidogenic peptide aggregation through photooxidation via1O2 generation.79 The same group also found that the fluorine-substituted 2-phenylquinoline-based solvento complex 6 showed good ability to interact with Aβ peptide in a coordination-dependent manner like previously reported solvento iridium(III) complexes as well as through photooxidation (Fig. 1).80 Very recently, Lee's group reported that a boron-dipyrromethene-based iridium(III) complex 7 can oxidize amyloidogenic peptides through both type I and type II processes.81 The introduction of the boron-dipyrromethene moiety overcomes the drawbacks of weak absorbance in the short wavelength region and transient triplet excited states, indicating that the combination of organic dyes and iridium(III) complexes can improve the PDT performance of the complexes.
Apart from histidine, solvento iridium(III) complexes have also been found to coordinate with glutamine (Gln). In 2016, Mao's group designed solvento iridium(III) complex 8 featuring an aldehyde group for the selective detection and imaging of Gln in live cells,82 but interactions between iridium(III) complexes and Gln are less studied (Table 1). Other examples of using iridium(III) complexes as photocatalysts for modifying bioactive peptides are out of the scope of this review.83,84 In the following sections, we focus on the conjugation of iridium(III) complexes with functional peptides to generate bifunctional iridium(III)–peptide bioconjugates, similar to the well-known ADC and proteolysis-targeting chimera (PROTAC) that have received significant attention for bioanalytical and biomedical applications in recent years.82
Complex | Target | Solvent | λ abs/nm | λ emi/nm | Φ PL | Φ Δ | Lifetime/ns | Ref. |
---|---|---|---|---|---|---|---|---|
1 (H2O/CH3CN) | Histidine/histidine-rich proteins | — | — | — | — | — | — | 69 |
2 | Aβ1–40 fibrils/Aβ1–40 monomers | — | — | — | — | — | — | 70 |
3 | Aβ1–40 fibrils/Aβ1–40 monomers | — | — | — | — | — | — | 70 |
4 | Aβ1–40 fibrils/Aβ1–40 monomers | PBS | 280, 422 | 586 | 0.0806 | — | 4.502 | 72 |
5 | Aβ peptides | HEPES | 463 (H2O) | 600 | 0.41 | 0.25 | 238 | 79 |
6 | Aβ peptides | H2O | 274, 336, 446 | 589 | 0.0071 | — | 4.8 | 80 |
7 | Aβ peptides | CH2Cl2 | 505 | 563 (H2O) | 0.1738 | — | — | 81 |
8 | Gln | DMSO/PBS | 281, 318, 420 | 557 | 0.04 | — | 100 | 82 |
Later on, Fei's group investigated the distance between histidines for cyclic formation.86 They found that two or three amino acids gave robust results, generating bioconjugates 10 and 11, respectively (Fig. 2). Based on this result, they developed the iridium(III) bioconjugate 12, containing the coordination-based cyclized functional peptide, cRGD, for cancer cell targeting. This method allows a convenient and convincing assessment of the cell selectivity of therapeutic peptide compounds. This work elegantly utilized iridium(III) coordination to rigidify peptide conformation, and amplify the biological activities of peptides.
Instead of coordination, the amide coupling strategy has been widely applied to generate iridium(III)–peptide conjugates. In 2011, Leeuwen's group synthesized peptide-functionalized bioconjugates 13–15 for the lifetime imaging of chemokine receptor 4 (CXCR4) expression using traditional amide bond formation strategies (Fig. 3).87 The CXCR4 is a G-protein-coupled membrane receptor, which plays an important role in tumor progression and metastasis, and is thus emerging as a diagnostic and targetable delivery target.88,89 The peptide Ac-Tz14011 has been identified as antagonist for CXCR4. Thus, 2-phenylpyridine-based triscyclometalated iridium(III) bioconjugates 13–15 bearing different numbers of Ac-Tz14011 were synthesized by forming an amide bond between the iridium(III) complex and the peptides. For the design of peptide–iridium(III) complex conjugates, this group initially investigated the crystal structure of CXCR4 complexed with the Ac-TZ14011 analogue to confirm that the luminescent iridium(III) label did not interfere with receptor binding, as the conjugation site in the peptide (d-Lys8) is distant from the pharmacophore. All three iridium(III)–peptide conjugates could visualize CXCR4 expression in CXCR4-transfected MDA-MB-231 cells (Fig. 3), and the trimeric conjugate 15 showed the best KD of 66.3 nM and ratio of MFIRs (mean fluorescence intensity ratio) of 1.72. The long lifetime of bioconjugate 14 was also used for fluorescence lifetime imaging (FLIM) of CXCR4. In the cell viability experiment, bioconjugate 13 displayed low toxicity at 10 μM, whereas 14 and 15 were considerably more toxic. The reduced viability of derivatives 14 (+10 charge) and 15 (+15 charge) could be the result of the high positive charge, compared with Ac-Tz14011 with +5 charge. This is the first example of a neutral iridium(III) complex functionalized with peptides for specific targeting of a cancer-associated membrane receptor, but its application in native CXCR4-expressing cancer cells was not explored.
![]() | ||
Fig. 3 (A) Chemical structure of bioconjugates 13–15. (B) Confocal microscopy and transmission images of the peptide-conjugated iridium(III) complexes on MDAMB231CXCR4+ cells: (A) 1 μM of 13; (B) 1 μM of 14; (C) 1 μM of 15. Reproduced with permission from ref. 87. Copyright 2011, John Wiley and Sons. |
Cationic cyclometalated iridium(III) complexes can be easily synthesized using a modular strategy.90–92 First, chloro-bridged iridium(III) dimer precursors are prepared via the reaction of cyclometalated C^N ligands and iridium(III) chloride, which are then reacted with the appropriate ancillary N^N ligand to prepare the iridium(III) complexes.93 Typically, the cyclometalated C^N ligands are used to modulate the photophysical properties of the bioconjugates, while ancillary N^N ligands are used to tether peptides.34 In 2018, Ma's group developed a permanent formyl peptide receptor 2 (FPR2) imaging probe 16, which conjugated a WKYMVm (FPR2 agonist) to the ancillary ligand of an iridium(III) complex (Fig. 4).94 Formyl-peptide receptors (FPRs) are therapeutic targets for a variety of diseases, including cancer, inflammation and neurodegenerative disease.95 Bioconjugate 16 was synthesized by a standard solution-phase peptide protection and deprotection strategy. Bioconjugate 16 showed a maximum emission at 576 nm with a maximum excitation at 291 nm in DMSO. The probe imaged FPR2-expressing HUVEC cells in a dose- and time-dependent manner, and selectively targeted FPR2 in HUVEC cells, and also inhibited lipoxin A4 (LXA4)-triggered cell migration in HUVEC cells, thereby serving as a theranostic probe.
In 2017, Marchán's group reported a bioconjugate 17 to a tumor-targeting vector based on octreotide (OCT) peptide, (D-Phe)-Cys-Phe-(D-Trp)-Lys-Thr-Cys-Thr (Fig. 4).96 The attachment of the iridium(III) complex to OCT was achieved through the formation of an amide bond between the carboxylic group in the benzimidazole diimine ligand and the N-terminal end of the peptide sequence through a stepwise solid-phase strategy. Bioconjugate 17 accumulated in cancer cells overexpressing somatostatin subtype-2 receptor (SSTR2), and the participation of the receptor was confirmed by competitive experiments. In cell imaging, bioconjugate 17 allowed the visualization of luminescent vesicles in the cytoplasm, most likely in endosomes, confirming the cellular uptake of the bioconjugate 17 in HeLa cells. Further imaging reflected a slightly lower cellular accumulation in MDA-MB-231 cells compared with that of HeLa cells as indicated by a reduced intensity of the luminescence signal. Overall, these results demonstrate that the design of iridium(III)–peptide conjugates can improve tumor selectivity. Moreover, this study highlighted the importance of a spacer (ethylene glycol flexible chain) to keep the metal complex away from the pharmacophore sequence and the β-turn peptide structure, which are key elements for recognition and binding to the receptor.
Ma's group also developed a gastrin-releasing peptide receptor (GRPr) imaging probe using a GRPr peptide antagonist, the statine-based JMV594 [(D-Phe)-Gln-Trp-Ala-Val-Gly-His-Sta-Leu-NH2] (Fig. 4).97 Bioconjugate 18 was synthesized through a standard solution-phase peptide protection and deprotection strategy. 6-Aminohexanoic acid was used as a long-chain linker to combine the iridium(III) complex with JMV594, in order to reduce the likelihood of the iridium(III) complex interfering with the recognition ability of JMV594. Bioconjugate 18 exhibits an intense absorption band at 250–310 nm in CH3CN and shows an excitation maximum at 321 nm and an emission maximum at 596 nm in DMSO, with a large Stokes shift of 275 nm. Bioconjugate 18 displayed negligible cytotoxicity against A549 cancer cells and the normal human hepatic cell line LO2 cells. In contrast, the iridium(III) complex without the peptide exhibited high toxicity against both A549 cells with an IC50 of 7.31 μM, and moderate toxicity against LO2 cells with of an IC50 of 27.75 μM. Thus, the incorporation of JMC594 abolishes the toxicity of the iridium(III) complex owing to reducing undesirable non-specific binding to biomolecules. Finally, bioconjugate 18 displayed negligible luminescence in living LO2 cells, but stronger emission in GRPr-expressing A549 cancer cells, which shows the potential of the probe for the diagnosis of cancer.
Organelle-specific probes are key to studying biological processes at the subcellular level.98,99 Iridium(III)–peptide bioconjugates have been developed for specific targeting of organelles via conjugation with suitable peptides. In 2012, Klimant's group developed bioconjugates 19–21 as cellular O2 sensors (Fig. 4).54 Attaching different short peptide sequences to the iridium(III)-octaethylporphyrin provided different targeting abilities via simple ligand-exchange reactions by histidine residues. Specifically, bioconjugate 19 possessed histidine-tetraarginine with +9 charge, bioconjugate 20 possessed a truncated fragment of cell-penetrating bactenecin 7 peptide, PRPLP, with +3 charge, while bioconjugate 21 carried the RGD sequence, a tumor-cell-targeting vector, with +1 charge. Absorption and emission spectra of the bioconjugates were found to be similar. The complexes showed a Soret band at around 386 nm, Q-bands at 506 and 540 nm, and peptide absorption in the UV region. Their emission was at around 654 nm and they also showed unquenched lifetimes above 40 μs in PBS. Both 19 and 20 demonstrated efficient staining of MEF cells, accumulating in the perinuclear regions and partially colocalizing with endoplasmic reticulum (ER) Tracker Green, a marker for the ER. Moreover, positive cytoplasmic staining was observed for all of these cell lines, including COS-7, HeLa, SH-SY5Y, and PC12 cells and mixed cultures of primary neurons and astrocytes. However, bioconjugate 21 carrying the RGD peptide unexpectedly did not display specific staining in HeLa and SH-SY5Y cells. This indicates that it is possible for metal conjugation to abolish the binding functionality of RGD peptides, possibly due to steric interference, the hydrophobicity of the porphyrin core, or the positive charge of the iridium(III) ion. As a result, when designing bioconjugates of this kind, it is crucial to consider the impact of the iridium(III) moiety on the activity of the peptide.
In 2019, Pope's group reported a cationic cyclometalated iridium(III)–peptide bioconjugate 22, which linked a nuclear localization signal (NLS) peptide (PAAKRVKLD) into the ancillary ligand (Fig. 4).100 To generate the PAAKRVKLD sequence, Fmoc-solid-phase peptide synthesis was used. The NLS peptide originates from human C-MYC protein and was first identified in 1988 to be essential for the nuclear localization of the protein, and three of its residues are cationic at physiological pH. Bioconjugate 22 had similar photophysical properties to the parent complex, which showed maximum emission at 677 nm in dilute aerated aqueous solvent and a lifetime of 38 ns. Imaging studies showed that incubation with 80–100 μM bioconjugate 22 promoted good cell uptake and nuclear localization in human fibroblast cells. In comparison, a structurally related, photophysically analogous iridium(III) complex lacking the peptide sequence showed very different biological behavior, with no evidence of nuclear, lysosomal or autophagic vesicle localization and significantly increased toxicity to the cells at concentrations >10 μM with induced mitochondrial dysfunction. Thus, it can be concluded that the NLS peptide successfully endows the translocation and nuclear uptake properties of 22, whilst lowering toxicity. This study demonstrates the use of subcellular targetable units to modulate the intracellular localization behavior of iridium(III) complexes.
In 2020, Lo's group also reported a organelle-targeting peptide-conjugated iridium(III) complex 23 through a perfluorobiphenyl (PFBP) clickable moiety, which is easily bioconjugatable with a thiol group for bioimaging and phototherapeutic applications (Fig. 5).101 Complex 23 showed a maximum emission at 577 nm with a maximum excitation at 350 nm, and a lifetime of 0.40 μs in PBS and CH3CN (1/1). Taking advantage of the high reactivity and unique chemo- and regioselectivity offered by the π-clamp sequence, the reaction between the PFBP complexes with cysteine-containing peptides containing the π-clamp sequence afforded luminescent conjugates with desirable photophysical and photochemical characteristics and differential cellular uptake and localization properties in HeLa cells. This strategy enables easy preparation of structurally diverse iridium(III)–peptide bioconjugate, and offers the possibility of the design of new theranostic agents.
![]() | ||
Fig. 5 Modification of PFBP compound with peptides through the π-clamp-mediated cysteine conjugation. Reproduced with permission from ref. 101. Copyright 2020, American Chemical Society. |
Bimetallic molecules offer the possibility of improving solubility and enhancing biological activities. In 2020, Gimeno's group reported a luminescent bimetallic iridium(III)/Au(I)–peptide bioconjugate 24 as a potential theranostic agent, which introduced an bioactive Au(I) complex onto the side chain of the peptide, Leu-enkephalin (Tyr-Gly-Gly-Phe-Leu), which binds to the opiate receptors in the central nervous system (Fig. 6).102 Some carcinomas, such as colorectal and pulmonary cancer, contain high levels of opioid peptides and their corresponding membrane-bound opioid receptors.103 Thus, linkage of drugs to opioid peptides could enhance cancer cell recognition and cellular uptake. The pentapeptide was synthesized by solid-phase peptide synthesis, then linked to the iridium(III) fragments via an amide bond using a carboxylic acid-functionalized diimine ligand. Finally, alkynyl groups were chosen as handles for insertion of the Au fragment. The bioconjugate 24 showed a maximum emission at 613 nm with a maximum excitation at 375 nm in DMSO. The probe 24 showed a red-shifted emission at around 615 nm, which was attributed to the electron-withdrawing amide substituent that stabilizes the π* orbitals of the diamine. However, this bioconjugate had no cytotoxicity against A549 cells, which is not common among Au(I) complexes, which was attributed to the inability of cationic complexes to escape the lysosome. In cell imaging, the bioconjugate 24 was not localized to the mitochondria in A549 cells, and localized in lysosomes with Pearson coefficients ranging from 0.8–0.9. This result was also unexpected, as some iridium(III)/Au(I) complexes generally have mitochondria targetability. Possibly, the presence of protonatable basic substituents could facilitate lysosomal localization instead. This work opens the door for the development of bimetallic peptide bioconjugates for biomedical applications, but their diagnostic potential was not explored.
Inspired by the aggregation of Aβ peptides, FF-rich peptides have self-assembling ability, and are widely applied for enhancing cell penetration, bioimaging and therapeutic performance.104 Recently, Chao's group developed a self-assembling bioconjugate 25 through conjugating the naphthalene-Phe-Phe-Lys moiety to C^N ligands, to achieve long-term lysosome imaging (Fig. 6).105 The triphenylamine group on the N^N ligand enabled bioconjugate 25 to display aggregation-induced emission (AIE) features, generating emission at around 600 nm in H2O. Further experiments found that bioconjugate 25 exhibited pH-responsive self-assembly due to π–π interactions among the naphthalene and phenylalanine units. This stronger interaction leads to regular-shaped nanoparticles at pH 7, while a weaker π–π stacking at pH 4 triggers the molecular self-assembly, forming irregularly shaped networks. Moreover, 25 successfully captured the lysosome dynamic response to mitochondria in HeLa cells through structured illumination microscopy (SIM).
Very recently, our group reported the first prostate-specific membrane antigen (PSMA) turn-on bioconjugate 26 (Fig. 6).106 Lys-urea-Glu, a PMSA inhibitor, was conjugated into iridium(III) complex.107,108 The bioconjugate 26 exhibited maximum excitation at 330 nm and a broad emission band with a maximum peak at around 670 nm in CH3CN. This peptide sequence exhibits excellent properties such as small size, high aqueous solubility, high affinity for PMSA (in the low nanomolar range), good biocompatibility and specificity. The luminescence emission lifetime of 26 was determined to be 24 ns (τ1) and 375 ns (τ2) in CH3CN. The binding affinity of 26 to PMSA was determined to be 9.49 nM, which is over 1000-fold lower than that of the positive control DUPA (12.4 μM). Moreover, 26 showed no obvious toxicity against the human prostatic carcinoma cell line LNCaP cells (IC50 > 50 μM) and the normal cell line LO2 (IC50 > 50 μM). In contrast, the parent complex without modification with Lys-urea-Glu was highly toxic to LNCaP cells (IC50 = 0.082 μM) and LO2 cells (IC50 = 0.50 μM). Furthermore, 22Rv1 tumors could be clearly visualized by probe 26 in a mouse xenograft model of prostate cancer. This is the first example of the in vivo use of an iridium(III)–peptide bioconjugate (Table 2).
Bioconjugate | Peptide | Target | λ abs/nm | λ emi/nm | Φ PL | Lifetime/ns | Ref. |
---|---|---|---|---|---|---|---|
9 | CPP/MTS | Cell-penetrating/cell-crossing | — | — | — | — | 85 |
10 | — | — | — | — | — | — | 86 |
11 | — | — | — | — | — | — | 86 |
12 | cRGD | Cancer cell | — | — | — | — | 86 |
13 | Ac-Tz14011 | CXCR4 | — | — | — | — | 87 |
14 | Ac-Tz14011 | CXCR4 | — | — | — | — | 87 |
15 | Ac-Tz14011 | CXCR4 | — | — | — | — | 87 |
16 | Gly-Met-Val-Met-Try-Lys-Trp | FPR2 | 255 (CH3CN) | 576 (DMSO) | — | 4620 | 94 |
17 | (D-Phe)-Cys-Phe-(D-Trp)-Lys-Thr-Cys-Thr | SSTR2 | — | — | — | — | 96 |
18 | (D-Phe)-Gln-Trp-Ala-Val-Gly-His-Sta-Leu | GRPr | — | 596 (DMSO) | — | 149 | 97 |
19 | Arg-Arg-Arg-Arg | Cell-penetrating | 386, 506, 540 (PBS) | 654 (PBS) | 0.13 (PBS) | 58 | 54 |
20 | Pro-Arg-Pro-Arg-Leu-Pro | Cell-penetrating | 388, 506, 540(PBS) | 652 (PBS) | 0.08 (PBS) | 69 | 54 |
21 | Gly-Arg-Gly-Asp | Cancer cell | 386, 507, 540(PBS) | 654 (PBS) | 0.13 (PBS) | 47 | 54 |
22 | Pro-Ala-Ala-Lys-Arg-Val-Lys-Lau-Asp | Nucleus | — | 677 (dilute aerated aqueous solvent) | — | 38 | 100 |
23 | — | — | — | 577 (PBS/CH3CN) | 0.13 (PBS/CH3CN) | 400 | 101 |
24 | Tyr-Gly-Gly-Phe-Leu | Opiate receptors | 266, 381, 416 (DMSO) | 613 (DMSO) | 0.021 (DMSO) | — | 102 |
25 | Phe-Phe-Lys | Self-assembly | — | 600 (H2O) | — | — | 105 |
26 | Lys-urea-Glu | PMSA | — | 670 (CH3CN) | 0.0211 (CH3CN) | 24, 375 | 106 |
Octaarginine also possesses CPP properties. In 2013, Keyes's group examined the conjugation of an iridium(III) luminophore to octaarginine (Fig. 7).114 The octaarginine peptide was prepared using a solid-phase peptide synthesizer. Unlike the parent complex, bioconjugate 28 was soluble in aqueous media and did not require pre-dissolution in an organic solvent. The UV–vis spectrum of bioconjugate 28 was dominated by a peptide absorbance peak at 262 nm an absorbance at 330 nm in PBS (pH 7.4). The MLCT absorbance peak at 387 nm was also apparent, whereas for the complex without the peptide, this band is broad, tailing to approximately 480 nm. The bioconjugate 28 exhibited a maximum emission at 543 nm in H2O, which is slightly red-shifted compared with the parent (537 nm). Notably, bioconjugate 28 was rapidly and irreversibly transported across the cell membrane of both SP2 and CHO cells at room temperature. Moreover, bioconjugate 28 showed substantially higher cytotoxicity in both cell lines than the parent complex. The IC50 values of 28 against SP2 and CHO cells were 34.95 μM and 54.44 μM, respectively, compared with 84.84 μM and 87.97 μM, respectively, for the parent iridium(III) complex. Overall, this work demonstrates the value of octaarginine conjugates for cell delivery of metal complexes.
Based on the iridium(III) coordination-based cRGD strategy, Fei's group also reported solvento iridium(III) complexes cyclized with oligoarginines as promoters of oncotic cell death (Fig. 7).115 The parent solvento complex or oligoarginine peptides alone showed no measurable toxicity (IC50 > 100 μM). Encouragingly, the cytotoxicity of the iridium(III)–peptide bioconjugates against HeLa cells was distinctly enhanced upon iridium(III) coordination (IC50 < 3.12 μM). Moreover, iridium(III) complexes with linear peptides were almost always less toxic than iridium(III) with cyclic peptide, especially for the peptides that have fewer numbers of arginines, possibly because of the higher hydrophobicity of linear peptide–iridium(III) complexes compared with their cyclic analogues. Cyclization of peptides could enhance their endocytosis, while the number of guanidine groups was related to their cell uptake efficiency. When the arginine number reached 8, both linear peptide–iridium(III) complexes and cyclic peptide–iridium(III) complexes showed similar cell uptake, suggesting that the flexibility of the linear structure may have been overwhelmed by a stronger electrostatic interaction between octaarginine and the cell membrane. The bioconjugates 29–30 with octaarginines were found to be most cytotoxic, and induced progressive oncotic cell death featuring cell membrane penetration and eruptive cytoplasmic content release. Experiments showed that the bioconjugate 30 utilized an energy-independent pathway to enter HeLa cells, where it distributed preferentially in the ER and the mitochondria rather than in the lysosomes. These bioconjugates can overcome multiple chemical drug resistances of cancer cells, and are immunogenic by stimulating dendritic cell maturation and inflammatory factor accumulation in mouse tumors.
In the targeted imaging studies, peptide conjugation of iridium(III) complexes lowered the cytotoxicity of the iridium(III)–peptide bioconjugates to normal cells, as shown by bioconjugates 18, 22 and 26. In a similar vein, targeted therapy is regarded as a promising strategy for overcoming the high cytotoxicity of metallodrugs due to poor selectivity against cancer cells.89,116 In 2015, Aoki's group reported a triscyclometalated bioconjugate 31 containing a cationic peptide as an inducer and detector of cell death via a calcium-dependent pathway (Fig. 8).117 KKGG (Lys-Lys-Gly-Gly) is a cationic amphiphilic peptide with +9 net charge which selectively interacts with negatively charged cancer cell membranes owing to electrostatic and hydrophobic interactions. The amphiphilic triscyclometalated iridium(III) complex bioconjugate was designed by tethering a KKGG peptide, which is an inducer and detector of cell death, via an alkyl chain linker to the cyclometalated ligand. The bioconjugate 31 showed a maximum emission at 509 nm with a maximum excitation at 366 nm in degassed 100 mM HEPES (pH 7.4). Its absorption peaks were at 280 nm and 362 nm, while its emission lifetime was 1.7 μs. The bioconjugate showed lower toxicity against normal mouse lymphocytes compared with cancer cells, such as Jurkat and HeLa S3 cells, due to the presence of the KKGG peptide. In Jurkat cells, the bioconjugate interacts with anionic molecules on the surface and membrane receptors to trigger the Ca2+-dependent pathway and intracellular Ca2+ response, resulting in necrosis accompanied by membrane disruption. This work shows the potential of amphiphilic iridium(III)–peptide bioconjugates in anticancer drug discovery, but the role played by the iridium(III) complex part was not further clarified in this study. This work provides an excellent example of water-soluble peptide–iridium(III) complex conjugates for biological and biomedicinal applications. In 2016, López's group synthesized a new family of cyclometalated iridium(III) oligocationic peptides 32–33 conjugated with three arginine residues.118 The cyclic peptides bioconjugates 32–33 displayed IC50 values in a similar range (19 and 21 μM, respectively) in NCI/ADR-RES cell lines, which are comparable to cisplatin (14 μM) (Fig. 8). Bioconjugates 32–33 localized and aggregated on the cell membrane, until they reached a threshold local concentration that allows them to behave like detergents and thus degrade the phospholipid bilayer causing cell death, which occurred on a much faster timescale than the toxicity effects of cisplatin.
Later on, Aoki's group further explored the Ca2+-related mechanism of this bioconjugate using proteomics, and introduced a photoaffinity unit, 3-trifluoromethyl-3-phenyldiazirine (TFPD), into at the end of the KKKGG peptide on the bioconjugate (Fig. 8).119 The protected TFPD-KKKGG peptide was prepared by Fmoc solid-phase peptide synthesis and was coupled with the amino group from the iridium(III) complex ligand, thereby producing bioconjugate 34 with +9 net charge. Bioconjugate 34 showed a maximum emission at 509 nm with a maximum excitation at 366 nm, and showed absorption peaks at 280 nm and 371 nm in degassed 100 mM HEPES. Three mechanisms were found: the first scenario involves the inhibition of (Ca2+-CaM)-PMCA complexation by iridium(III) complexes near the plasma membrane, thus preventing channel opening; the second involves the inhibition of (Ca2+-CaM)-K-Ras4B complexation by bioconjugate 34 to facilitate the phosphorylation of K-Ras4B in Jurkat cells; the third is the activation of the G-protein transduction pathway by the iridium(III) complexes, resulting in the release of Ca2+ from the ER to the cytoplasm in Jurkat cells. Overall, these mechanisms are associated with intracellular Ca2+ overload, thus inducing cell death.
In 2017, the same group further optimized the peptide substitution position with cationic peptides at the 4′ position of the 2-phenylpyridine ligand (Fig. 8).120 The MTT experiment demonstrated that the 4′-KKGG ppy ligand had higher cytotoxicity compared with the 5′-KKGG ppy ligand in Jurkat cells. Moreover, the linker length is also key to their cytotoxicity; the eight-carbon alkyl length of bioconjugate 35 (EC50 = 2.4 μM) on the 4′-ppy ligand had stronger cytotoxicity compared with other complexes with different linker lengths (29 μM for two-carbon alkyl length, 5.0 μM for six-carbon alkyl length, and 34 μM for sixteen-carbon alkyl length). These cytotoxicity results caused by linker length are consistent with previous studies, indicating that a balance between the hydrophobicity and hydrophilicity of the iridium(III) complex is an important factor. The bioconjugate 35 showed a maximum emission at 541 nm with a maximum excitation at 366 nm with a lifetime of 1.0 μs, and showed absorption peaks at 283 nm and 383 nm in degassed 100 mM HEPES. This work indicates that we need to carefully choose the connection position and linker length when designing iridium(III)–peptide bioconjugates. Later on, the group connected the KKKGG peptide through a nucleophilic substitution reaction on the 2-phenylpyridine ligand instead of through amide bond formation, to give the bioconjugate 36, with +12 net charge (Fig. 8).121 Bioconjugate 36 showed a maximum emission at 505 nm with a maximum excitation at 366 nm, with a lifetime of 0.78 μs. It also showed absorption peaks at 292 nm and 356 nm in degassed 100 mM HEPES. Bioconjugate 36 could induce paraptosis-like cell death of cancer cells via an intracellular Ca2+-dependent pathway in Jurkat cells with an EC50 of 1.5 μM.
Aoki's group further introduced a hydrophobic N-heptadecanoyl group at the N-terminus of the KKKGG peptide to obtain the bioconjugate 37 (Fig. 8).122 The bioconjugate 37 showed a maximum emission at 501 nm with a maximum excitation at 365 nm, with a lifetime of 1.4 μs. Its absorption peaks were at 284 nm and 366 nm in degassed 100 mM HEPES. The MTT cytotoxicity assay showed an EC50 of 1.5 μM against Jurkat cells, while 37 weakly induced cell death of IMR90 cells. It was found that bioconjugate 37 accumulated on the cell membrane and induced cell death via affecting the mitochondrial Ca2+-dependent pathway and an intracellular Ca2+ response. This study demonstrates that the appropriate choice of alkyl structure at the N-terminus in the peptide units could be used to tailor the lipophilicity of iridium(III)–peptide bioconjugates, which may enhance anticancer activity and improve cancer cell selectivity.
Necrosis and apoptosis are the two main pathways to cell death.123,124 In 2018, Aoki's group designed and synthesized luminescent iridium(III)–peptide bioconjugate 38 for imaging cancer cells and inducing apoptosis (Fig. 9).125 Bioconjugate 38 showed a maximum emission at 506 nm with a maximum excitation at 366 nm with a lifetime of 1000 ns, and showed absorption peaks at 286 nm and 360 nm in degassed DMSO at 25 °C. The hydrophilic SGSG sequence was inserted between the iridium(III) complex core and the cyclic peptide sequence WDCLDNRIGRRQCVKL as a tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) to improve the complex's solubility. Upon binding of TRAIL with death receptor 5 (DR5), this cyclic peptide sequence can interact with DR5 to induce cancer cytotoxicity dependent on the DR5 expression level of Jurkat cells. Bioconjugate 38 is the first example of an artificial luminescent TRAIL mimic that induces apoptosis-like cell death, although its cytotoxicity was only moderate.
Soon after, Aoki's group designed and synthesized another iridium(III)–peptide bioconjugate 39 similar to 38 to image cancer cells and induce necrosis-type cell death (Fig. 9).126 Bioconjugate 39 showed a maximum emission at 506 nm with a maximum excitation at 366 nm with a lifetime of 1.3 μs, and showed absorption peaks at 285 nm and 361 nm in degassed DMSO. They found that cancer cell cytotoxicity of bioconjugate 39 was also dependent on the DR5 expression level of Jurkat cells, but bioconjugate 39 induced cell death via necrosis, as opposed to apoptosis for bioconjugate 38. The cell death induced by 39 was considered a necrosis-type cell death via a Ca2+-mediated intracellular signaling pathway.
In 2021, Aoki's group designed iridium(III) bioconjugates 40–42, containing three, two or one KKKGG sequences, respectively, to assess the effect of the number of peptide units on anticancer activity (Fig. 9).127 Bioconjugates 40–42 exhibited a maximum emission at 507, 502 and 498 nm in degassed 100 mM HEPES and the emission quantum yields were determined to be 0.41, 0.44 and 0.12 in degassed 100 mM HEPES, respectively, and their lifetimes were determined to be 1.4, 1.6 and 1.8 μs, respectively. The EC50 values of bioconjugates 40–42 against Jurkat cells were determined to be 1.5 μM, 4.1 μM, and 16.3 μM, respectively. These data show that there is a positive correlation between the number of H2N-KKKGG peptide groups and the degree of cytotoxicity against Jurkat cells. Meanwhile, the cytotoxicity of bioconjugates 40–42 was relatively lower against other cell lines, including cancer cells HeLa S3 (EC50 > 50 μM) and A549 cells (EC50 > 39.3 μM), and human diploid cells IMR90 cells (EC50 > 18.4 μM). Recently, the same group discovered that these bioconjugates could induce cell death through new pathways: (1) by triggering paraptosis in cancer cells via an intracellular Ca2+ overload from the ER and a decrease in mitochondrial membrane potential,128 and (2) mitochondrial Ca2+ overload triggered by membrane fusion between mitochondria and the ER.129
Click chemistry is a versatile methodology for establishing structurally diverse iridium(III)–peptide bioconjugates. Cu(I)-catalyzed azide–alkyne cycloadditions have been a benchmark in many areas of recent synthetic chemistry; however, side reactions and toxicity of the copper catalyst often limits its utilization in biological applications. By replacing terminal alkynes with strain-promoted cycloalkynes, copper-free azide–alkyne cycloadditions can be achieved under mild conditions without disrupting the function of the biomolecules. In 2019, Sadler's group reported a strategy for conjugating an iridium(III) complex to target peptides via copper-free click chemistry.130 They synthesized a half-sandwich Cp* iridium(III) complex with a dibenzocyclooctyne moiety 43, which showed moderate cytotoxicity against human ovarian cancer cells A2780. The center of the iridium(III) complex can catalyze hydride transfer from NADH to molecular oxygen, generating hydrogen peroxide in cancer cells to trigger cell death. However, it lacked selectivity against tumor cells over normal cells. Subsequently, complex 43 was conjugated with a tumor-targeting cyclic nonapeptide c(CRWYDENAC) to afford bioconjugate 44 (Fig. 9). Bioconjugate 44 had enhanced anticancer activity and selectivity due to the introduction of the peptide (Table 3).
Bioconjugate | Peptide | Target | λ abs/nm | λ emi/nm | Φ PL | Lifetime/ns | Ref. |
---|---|---|---|---|---|---|---|
27 | Octa-arginine | DNA | — | — | — | — | 113 |
28 | Octa-arginine | Cell-penetrating | 262, 330, 387(PBS) | 543 (H2O) | — | — | 114 |
29 | Octa-arginine | Cell-penetrating | — | — | — | — | 115 |
30 | Octa-arginine | Cell-penetrating | — | — | — | — | 115 |
31 | Lys-Lys-Gly-Gly | Cell surface/membrane receptors | 280, 362 (HEPES) | 509 (HEPES) | 0.55 (0.1 M H2SO4) | 1700 | 117 |
32 | Arg-Arg-Arg | — | — | — | — | 118 | |
33 | Arg-Arg-Arg | — | — | — | — | 118 | |
34 | Lys-Lys-Lys-Gly-Gly | Cell surface/membrane receptors | 280, 371 (HEPES) | 498, 509 (HEPES) | 0.0024/0.0062 (0.1 M H2SO4) | 119 | |
35 | Lys-Lys-Gly-Gly | Cell surface/membrane receptors | 283, 383 (HEPES) | 541 (HEPES) | 0.11 (0.1 M H2SO4) | 1000 | 120 |
36 | Lys-Lys-Lys-Gly-Gly | Cell surface/membrane receptors | 292, 356 (HEPES) | 505 (HEPES) | 0.43 (0.1 M H2SO4) | 780 | 121 |
37 | Lys-Lys-Lys-Gly-Gly | Cell surface/membrane receptors | 284, 366 (HEPES) | 501 (HEPES) | 0.52 (0.1 M H2SO4) | 1400 | 122 |
38 | DR-binding peptide | DR5 | 286, 360 (degassed DMSO) | 506 (degassed DMSO) | 0.35 (0.1 M H2SO4) | 1000 | 125 |
39 | DR-binding peptide | DR5 | 285, 361 (degassed DMSO) | 506 (degassed DMSO) | 0.33 (0.1 M H2SO4) | 1300 | 126 |
40 | Lys-Lys-Lys-Gly-Gly | Cell surface/membrane receptors | 280, 360 (degassed HEPES) | 507 (degassed HEPES) | 0.50 (0.1 M H2SO4) | 1400 | 127 |
41 | Lys-Lys-Lys-Gly-Gly | Cell surface/membrane receptors | 277, 358 (degassed HEPES) | 502 (degassed HEPES) | 0.41 (0.1 M H2SO4) | 1600 | 127 |
42 | Lys-Lys-Lys-Gly-Gly | Cell surface/membrane receptors | 278, 352 (degassed HEPES) | 498 (degassed HEPES) | 0.12 (0.1 M H2SO4) | 1800 | 127 |
43 | — | — | — | — | — | — | 130 |
44 | c(Cys-Arg-Typ-Tyr-Asp-Glu-Asn-Ala-Cys) | Integrin α6 receptor | — | — | — | — | 130 |
In the future, there are some directions for accelerating the clinical transition of iridium(III)–peptide bioconjugates. NIR iridium(III)–peptide bioconjugates with stronger absorption and emission over 500 nm will be preferred due to their better tissue penetration. The cyclometalated ligand plays an important role in modulating emission and absorption properties. 2-Phenylquinoxaline has been demonstrated to be a relatively simple cyclometalated ligand for enabling NIR emission in iridium(III) complexes,131 while fluorophores are also good choices as ligands.132 Photoacoustic and ultrasound-induced luminescence abilities of iridium(III)–peptide bioconjugates could also be investigated, as these techniques have better penetration in vivo.133,134 The in vivo therapeutic potential of iridium(III)–peptide bioconjugates must be systematically evaluated, while their anticancer mechanisms also need to be clarified. In addition, the therapeutic application of iridium(III)–peptide bioconjugates in other diseases apart from cancer is often overlooked, but is worth exploring due to the known abilities of iridium(III) complexes to reduce liver injury and promote diabetic wound healing, among other applications.135,136 Furthermore, based on the advances in the PDCs, the development of new G protein-coupled receptor-binding peptides would be also desirable to advance iridium–peptide conjugates to the level of ADCs. With further investment, we believe that iridium(III)–peptide bioconjugates can make a significant impact in biomedical science, similar to the success of other conjugate strategies such as ADC and PROTAC.
Footnote |
† These authors contributed equally to this work. |
This journal is © the Partner Organisations 2024 |